The present invention relates to a hearing aid system for the in-situ fitting of hearing aids.
For persons with a hearing loss, the sensitivity of the ear will often be frequency dependent within the usual audible range, ie. the person may have almost normal sensitivity at certain frequencies, but a low sensitivity at others.
Since the object of the hearing aid is to give normal hearing at all frequencies, the amplification provided by the hearing aid must as a result also be frequency dependent, with a high amplification at frequencies where hearing sensitivity is low and zero or low amplification where hearing is normal or close to normal.
Because hearing losses vary from person to person the frequency dependency or amplification characteristic for the hearing aid should be adjustable, so that the hearing aid can be fitted to the actual hearing loss of the person.
One way is to separately measure an audiogram for the patient, ie. measuring sensitivity of the ear to different frequencies and sound pressures, using a test signal generator and a headphone, and adjust the settings of the hearing aid accordingly based on the audiogram.
Another way is the in-situ fitting where the audiogram is measured with the hearing aid placed in the ear and acting as an audio signal source instead of the headphone. This is described in eg. U.S. Pat. No. 5,710,819.
In the in-situ fitting procedure the hearing aid is coupled to an external control device, with which a generation of test signals for the receiver, ie. the output transducer of the hearing aid can be activated. The test signals may either be generated in the control device and delivered to the hearing aid, or they may be generated in the hearing aid in accordance with control signals from the control device. In both cases the built-in amplifier of the hearing aid is used to achieve the different levels for the test signals, and hence the output sound levels from the receiver. The control device further may further provide the power for the hearing aid during the fitting procedure.
Even though the use of the hearing aid itself in the fitting procedure has advantages, such as higher accuracy in the fitting of the frequency characteristic compared to the fitting using a separate audiogram, it does have some drawbacks.
One major drawback is that a very high dynamic output range for the acoustic test signals is needed for the fitting procedure.
This dynamic range is expressed as the difference between the maximum output level achievable and the inherent noise level in the amplifier.
The reason that this very high dynamic range is needed is that the amplifier on one hand should be able to deliver signals powerful enough to make the sounds output by the receiver exceed the hearing threshold for persons with severe hearing losses, eg. above 130 dB SPL (Sound Pressure Level). On the other hand, when measuring on persons with normal hearing in at least certain frequency ranges very low sound output levels are needed, and in such cases the inherent amplifier noise should not exceed the level of the test signal. The latter requiring that the amplifier noise does not exceed approximately 10 dB SPL.
Hence, the necessary dynamic range of the amplifier should exceed 120 dB if the hearing aid is to be fitted in-situ on any person with an unspecified hearing loss.
In fact, if the same amplifier is to be used in different hearing aids of different construction, in particular with different receivers having different responses, the dynamic range should be even higher, eg. 140 dB.
This dynamic range of 140 dB is far more than the dynamic range of 60-80 dB needed under normal circumstances when the hearing aid is used.
Achieving these high dynamic ranges is complex and costly in hardware, and would increase the costs of the amplifier and thus of the hearing aid, whereas lower dynamic ranges of say 90 to 100 dB are readily achieved with both analogue and digital amplifiers. For instance this higher dynamic range would normally in digital hearing aids require a higher number of bits to achieve the higher resolution.
From U.S. Pat. No. 3,818,149 and U.S. Pat. No. 5,321,758 it is known to attenuate the output signal from the final stage in analogue amplifiers by means of resistor components. However, none of these hearing aids are adapted for in-situ fitting, and hence do not have a need for the mentioned large dynamic range.
In U.S. Pat. No. 3,818,149 the attenuation of the analogue signal is done for the purpose of volume control by means of a voltage divider in the form of an adjustable potentiometer. Having such a voltage divider as the final stage before the receiver leads to increased power consumption. Power consumption is an important issue in hearing aids, in particular because these of aesthetic reasons are small, leaving little room for batteries. Having such a voltage divider in the output circuit of a hearing aid is therefore undesirable.
In U.S. Pat. No. 5,321,758 is described a programmable analogue hearing aid with multiple frequency bands. When the hearing aid is fitted, the various frequency bands may be attenuated individually. The sum of these individual frequency bands are amplified in an analogue output stage. For the purpose of achieving a desired overall gain of the hearing aid the analogue output signal from the output stage may also be attenuated. This last attenuation is fixed once in the fitting procedure for the hearing aid, and is not changed, unless the hearing aid is fitted anew. This attenuation is achieved by means of a number of resistors which may be connected in parallel with each others between the output of the amplifier and the receiver, ie. in series with the impedance of receiver. The receiver may also be connected directly to the output of the amplifier by short circuiting of all the resistors. Apart from the fact that this way of attenuation also incurs losses, it is further undesirable because the output characteristic of the receiver compared to a solution using a voltage divider will be more dependent on the impedance of the receiver, which may not be linear but depend on frequency.
Contrary to the above mentioned analogue amplifiers digital amplifiers, known as class D or switch mode amplifiers, may, in principle, be made practically loss free. They are therefore often used where there is a need for high efficiency of the amplifier, eg. in battery powered hearing aids. In such amplifiers a fixed voltage level is switched in pulses. The impedance of the receiver receives the full supply voltage during these pulses, giving rise to a current. To achieve a specific output signal the pulses are modulated to give a mean current corresponding to the desired signal. Because the output level may be regulated entirely by adapting the switching cycles there it has never been suggested to use voltage dividers in connection with digital amplifiers as this would compromise the desired high efficiency of the amplifier.